JOURNAL OF THE Vol. 40, No. 4WORLD AQUACULTURE SOCIETY August, 2009

Effect of Different Dietary Protein and Lipid Levels on GrowthPerformance and Body Composition of Juvenile Southern

Flounder, Paralichthys lethostigma, Reared in a RecirculatingAquaculture System

Md. Shah Alam and Wade O. Watanabe1

University of North Carolina Wilmington, Center for Marine Science, Aquaculture Program,601 South College Road Wilmington, NC, 28403-5927 USA

Harry V. DanielsNorth Carolina State University, Department of Zoology, 248 David Clark Labs, Raleigh, NC

27695 USA

Abstract

The effects of six formulated diets containing different protein and lipid levels on growthperformance and body composition of juvenile southern flounder were evaluated. Test diets wereprepared with a combination of three crude protein (CP) levels (45, 50 and 55%) and two crudelipid (CL) levels (10 and 15%). Diets (CP/CL) were as follows: 45/10, 45/15, 50/10, 50/15, 55/10, 55/15and a commercial diet (50/15). Southern flounder (1.10 g) were fed the respective diets for 42 d intriplicate recirculating tanks (20 fish/tank). Percent body weight gain (BWG) for fish fed diet 45/10(413%) and the commercial diet (426%) were significantly (P < 0.05) lower than fish fed other diets(823–837%). Increasing protein level from 45 to 50% produced a significant increase in BWG for the10% lipid diet (823%) but further increasing protein did not produce a significant effect on BWGirrespective of dietary lipid levels. Specific growth rate (SGR), feed intake, feed conversion efficiency(FCE), protein efficiency ratio (PER), and total lipid content in the whole body were significantlyaffected by different dietary protein and lipid levels. Results indicated that a combination of 50%protein and 10% lipid was optimal for the growth performance of southern flounder juveniles.

The southern flounder, Paralichthys letho-stigma (P. lethostigma), is a flatfish of thefamily Bothidae. It can be found in coastalwaters from Albemarle Sound, North Car-olina, through South Atlantic states to CorpusChristi Pass, Texas, with the exception of SouthFlorida (Wenner et al. 1990). This species cangrow well in seawater or in freshwater. Theireuryhaline character and tolerance of a widerange of temperatures make southern flounderan ideal candidate for aquaculture. The devel-opment of intensive culture methods for south-ern flounder is of great interest because of itsstatus as a highly desirable food and recre-ational species and its potential for commercialculture. Its landings have declined, leading tointerest in culturing native flatfishes for stockenhancement or food fish production. Various

1 Corresponding author.

institutions in North Carolina are devoting sub-stantial resources to address the major problemsassociated with the mariculture of this species,particularly in recirculating aquaculture sys-tem (Daniels and Watanabe 2003). The method-ology of spawning and larval rearing is welldocumented (Berlinsky et al. 1996; Watanabeet al. 2001; Daniels and Watanabe 2003; Watan-abe et al. 2006). One of the factors limitingcommercial aquaculture production of southernflounder is the limited researches regarding thenutritional requirements (Alam and Watanabe2005; Gao et al. 2005; Gonzalez et al. 2005).

Feed is one of the highest costs in theoperation of an aquaculture enterprise, and inmany cases reaches 50% of total expenses.Protein influences the economics of a farmingindustry by determining the feed cost (NRC1993). Lipid is the source of energy for fish, andif lipid level is not sufficient in the diet as an

© Copyright by the World Aquaculture Society 2009

513

514 ALAM ET AL.

energy source, protein is used as compensatoryenergy source. Therefore, dietary protein andlipid levels must be in balance for maximumgrowth of fish (Kim et al. 2004). It is importantto provide an adequate level and ratio ofprotein, lipid, and carbohydrate in diets in orderto reduce catabolism of protein for energy. Theprotein-sparing effects by lipid or carbohydratelevels in diets have been reported in some fishspecies (Cho and Kaushik 1990; De Silver et al.1991; Lee et al. 2002).

Although the dietary protein requirement forsouthern flounder had been studied (Gao et al.2005), no information is available on utilizationof lipid as an energy source at different dietaryprotein levels. The objective of this study wasto determine the effects of various levels oflipid and protein in diets on growth and bodycomposition of juvenile southern flounder.

Materials and Methods

Experimental Animal

Adult broodstock southern flounder held inphotothermally controlled tanks at the Uni-versity of North Carolina Wilmington Centerfor Marine Science (UNCW-CMS) Aquacul-ture Facility, Wrightsville Beach, North Car-olina were induced to spawn using luteinizinghormone releasing hormone analog (LHRHa)(Berlinsky et al. 1996; Watanabe et al. 2001;Watanabe et al. 2006). The fertilized eggs werehatched and reared through 70-d post hatch-ing at the North Carolina State University fishhatchery (Raleigh, NC) as described by Danielsand Watanabe (2003) and then transferred toUNCW-CMS for experimental studies. Prior tothe feeding experiment, fish were conditionedfor 2 wk to a commercial pellet diet (Skret-ting, Canada, 1.0 mm pellet, crude protein [CP]50%, and lipid 15%).

Experimental Diets

To study the effects of dietary protein andlipid on southern flounder, a 3 × 2 factorialexperiment was designed. Six experimentaldiets were formulated to contain three protein(45, 50, and 55%) and two lipid (10 and 15%)

levels. CP and crude lipid (CL) in the diets wereas follows: 45/10, 45/15, 50/10, 50/15, 55/10,and 55/15 (Table 1). In addition, a domesticgeneric commercial diet (50% CP and 15%CL) was compared with the formulated practi-cal diets. The protein sources of the formulateddiets were menhaden meal, squid meal, andkrill meal. Menhaden oil and soybean lecithinwere used as lipid sources. Menhaden mealwas increased to increase the protein level indiet while menhaden oil was increased to pro-vide a higher lipid level in the diets. Wheatstarch was used as a carbohydrate source toincrease energy level. All other ingredientswere formulated according to the recent stud-ies on nutrient requirements of other marineflatfish (Alam et al. 2002, 2003). All ingredi-ents were purchased locally except vitamin andmineral premix (Kohkin Chemical Co., Ltd,Kagoshima, Japan, Kagoshima University) andvitamin and mineral for marine fish. Proteinsources were sieved through a 500-μm meshsieve. Energy levels of the diets were calcu-lated based on 16.7, 37.7, and 16.7 kJ/g forprotein, lipid, and nitrogen-free extract, respec-tively (Garling and Wilson 1976). Diets wereprepared as described previously (Alam et al.2007), with some modifications. In brief, alldry ingredients were mixed with a feed mixer(Kitchen Aid Inc., St. Joseph, MI, USA), andthen to this, previously mixed menhaden oil andlecithin were added. Approximately 35–40% ofdistilled water was added to the ingredient mix-ture to facilitate pelleting by a meat chopper(model MIN0012, Jacobi-Lewis Co., Wilming-ton, NC, USA). After pelleting, diets were driedat 70 C in a constant temperature oven (DKM600, Yamato Scientific Co., Ltd., Japan) andthen stored at –20 C. The proximate compo-sition of the diets was analyzed (Table 2).

Experimental Conditions

To begin the experiment, southern floun-der (approximately 1.10 g initial weight) wereplaced in triplicate recirculating tanks (75 L, 20fish/tank). The recirculating aquaculture systemincluded a Kaldness moving beds-based (AnoxKaldness Inc., Providence, RI, USA) biofilter, a

EFFECT OF DIFFERENT DIETARY PROTEIN AND LIPID LEVELS 515

Table 1. Composition of diets.

Test diets g/100 g

Protein/lipid 45/10 45/15 50/10 50/15 55/10 55/15 Commercial 50/15

Menhaden meala 39 39 47 47 54.5 54.5Squid meala 20 20 20 20 20 20Krill meala 5 5 5 5 5 5Menhaden oila 2 7.5 1.5 6.5 1 5Soybean lecithinb 1.5 1.5 1.5 1.5 1.5 1.5Vitamin mixc 4 4 4 4 4 4Mineral mixc 4 4 4 4 4 4Wheat-starchd 18.5 13 11 6 4 0Glutend 5 5 5 5 5 5Attractantse 1 1 1 1 1 1Total 100 100 100 100 100 100Gross energy (GE) (calculated, kJ/g diet) 14.0 15.5 14.1 15.1 14.0 14.5P/E ratio (mg protein/kJ GE) 32.1 29.0 35.4 33.1 39.2 37.9

aInternational Proteins Corporation, St. Paul, MN.bADM Company, Decatur, IL.cKohkin Chemical Co., Ltd, Kagoshima, Japan (Alam et al. 2008).dSigma-Aldrich, St. Louis, MO.eAttractants; alanine, betaine, glycine, and taurine (0.25% of each).

Table 2. Proximate compositions of diets. Values areaverage of triplicate analysis (N = 3). (Commercial dietcontained 50% crude protein, 15% lipid, 8% ash, 10%moisture).

45/10 45/15 50/10 50/15 55/10 55/15

Protein 45.5 45.9 50.2 51.0 54.7 54.9Lipid 9.9 15.4 10.2 15.2 9.9 14.5Moisture 10.1 11.2 10.3 12.4 10.5 12.3Ash 10.2 11.0 11.4 12.5 12.9 13.6

bead filter (Aquaculture Systems Technologies,LLC, New Orleans, LA) to remove solids, aprotein skimmers for removal of small partic-ulate and dissolved materials and an ultravio-let (UV) sterilizer for disinfection. Fish werefed twice/d 0900 and 1600 h as much as theycould consume during a 20-min period, andthe amount of diet consumed was recordeddaily. Each tank was covered with a lid toprevent fish from jumping out. A 10:14 h light-dark photoperiod was maintained. Water qualitywas checked twice weekly. Water temperatureranged from 19.1 to 20.5 C during the feedingperiod and dissolved oxygen (6.5–7.8 mg/L)was maintained near saturation with continu-ous aeration. Ranges of other parameters werepH 7.7 to 8.0 and salinity 33.5 to 34.4 ppt. Fish

were weighed every 2 wks. Growth study wasconducted for a period of 6 wks. A pooled sam-ple of 10 fish at the beginning and 10 fish pertank at the end of the feeding trial were sac-rificed and used to determine the whole bodyproximate composition.

Proximate Composition and StatisticalAnalyses

CP (nitrogen combustion) and crude fat con-tent (ether extraction) and moisture and ashcontent of the diets and fish whole body weredetermined using AOAC (1990) method atNew Jersey Feed Laboratory Inc., Trenton, NewJersey. All data were subjected to statistical ver-ification using one way and two-way ANOVA(JMP, version 6.0, SAS Institute Inc., Cary, NC,USA). Significant differences between meanswere evaluated by Tukey Kramer test (Kramer1956). Probabilities of P < 0.05 were consid-ered significant.

Results

Effects on Growth Performance

After 14 d no significant (P > 0.05) dif-ferences in mean final body weight (FBW)

516 ALAM ET AL.

Feeding periods (d)

Fin

al b

ody

wei

ght

(g)

00

2

4

6

8

10

12

45/1045/1550/1050/1555/1055/15Commercial

14 28 42

Figure 1. Effect of different protein and lipid levels onfinal body weight (g) after 14, 28, and 42 d feeding trial.Values are means of triplicate tanks.

among treatments were observed (Fig. 1). After28 d, FBW among the juveniles fed 45/10 and45/15 and the commercial diet were signifi-cantly (P < 0.05) lower than those fed 50/10,50/15, 55/10, and 55/15 and similar trendswere observed after 42 d (Fig. 1). The per-cent body weight gain (BWG), specific growthrate (SGR), feed intake (FI), feed conver-sion efficiency (FCE), protein efficiency ratio(PER), and percent survival (SR) are pre-sented in Table 3. The lowest BWG (413%)

was found for juveniles fed the 45/10 diet.At this protein level, increasing lipid levelsfrom 10 to 15% significantly increased BWG(545%). Increasing protein levels from 45 to50% increased BWG at both lipid levels to 823and 830% respectively. Further increasing pro-tein level from 50 to 55% did not significantlyincrease BWG irrespective of lipid levels. After42 d, a significant (P < 0.01) interactive effectbetween dietary protein and lipid levels onBWG was observed (Table 3). SGR showedsimilar trends as observed for BWG data withlowest SGR at 45/10 and 45/15, but with nosignificant difference among the diets 50/10,50/15, 55/10, and 55/15. FI was not signif-icantly different among fish fed diets 45/10and 45/15; however, FI was significantly higherfor the juveniles fed 50/15, 55/10, and 55/15(7.49–7.83). FCE showed the same trends asBWG. FCE was significantly lower (P < 0.05)on the 45% protein diets than on 50 or 55%protein. At 45% protein, FCE was signifi-cantly (P < 0.05) higher at 45/15 (0.97) thanat 45/10 (0.76). However, there were no sig-nificant differences among 50/10, 50/15, 55/10,and 55/15 (1.17–1.29). The lowest FCE (0.76)was observed for fish fed 45/10 with no signifi-cant difference from the commercial diet (0.87).

Table 3. Body weight gains (BWG), feed intake (FI), feed conversion efficiency (FCE), percent survival (SR), specificgrowth rate (SGR), and protein efficiency ratio (PER) of juvenile southern flounder fed diets with different levels of crudeprotein and crude lipids (CP/CL) for 42 d. Values are mean ± SEM for triplicate tanks. Means with different letters in thesame column differ significantly (P < 0.05).

Diets (CP/CL) BWG (%) FI g/fish/42 d FCE SR (%) SGR %/d PER

45/10 413 ± 12.7c 6.16 ± 0.18bc 0.76 ± 0.04c 94 ± 0.33a 3.88 ± 0.06c 1.51 ± 0.08c

45/15 545 ± 21.6b 6.26 ± 0.22bc 0.97 ± 0.03b 96 ± 1.00a 4.41 ± 0.08bc 1.94 ± 0.05b

50/10 823 ± 20.8a 7.13 ± 0.30ab 1.29 ± 0.03a 97 ± 0.05a 5.26 ± 0.06a 2.31 ± 0.05a

50/15 830 ± 32.9a 7.49 ± 0.23a 1.25 ± 0.06a 94 ± 1.15a 4.96 ± 0.27ab 2.25 ± 0.06ab

55/10 837 ± 34.8a 7.56 ± 0.16a 1.23 ± 0.02a 98 ± 1.15a 5.33 ± 0.09a 2.01 ± 0.09ab

55/15 834 ± 12.9a 7.83 ± 0.20a 1.17 ± 0.05a 98 ± 1.33a 5.31 ± 0.03a 1.91 ± 0.02b

Com 426 ± 36.3c 5.63 ± 0.25c 0.87 ± 0.03c 95 ± 1.52a 3.91 ± 0.17c 1.53 ± 0.09c

Two-way ANOVA(P -value)

Protein (A) 0.0001 0.0001 0.0001 0.0265 0.001 0.0003Lipid (B) 0.0851 0.1739 0.3702 0.8128 0.5811 0.2898A × B 0.0297 0.8417 0.0037 0.0835 0.0203 0.0021

BWG (%) = ([final wet weight − initial wet weight]/initial wet weight) × 100.SGR = (ln [mean final weight] − ln [mean initial weight]/42 d) × 100.FCE = Total feed intake (g)/wet weight gain (g).PER = weight gain (g)/total protein intake in dry basis (g).

EFFECT OF DIFFERENT DIETARY PROTEIN AND LIPID LEVELS 517

PER also showed similar trends to BWG. PERwas generally lower on the 45% protein dietsthan on 50 or 55% protein. At 45% protein,PER was significantly (P < 0.05) higher at45/15 (1.94) than at 45/10 (1.51), while therewere no significant differences among 50/10,50/15, and 55/10 (2.01–2.31). PER in the 55/15(1.91) diet was not significantly different from45/15 (1.94). PERs were significantly lower forthe 45/10 (1.51) and commercial diet (1.53)than for the other diets. After the feeding trial, ahighly significant (P < 0.01) interactive effectbetween dietary protein and lipid levels on feedconversion ratio (FCE) and PER was observedbut not for FI (Table 3). Survival (94 to 98%)was not significantly different among treat-ments.

Effects on Body Proximate Compositions

Whole body moisture (77.8–78.1%) andprotein content (15.7–16.6) were not signif-icantly different among treatments (Table 4).Body lipid content was significantly (P < 0.05)affected by dietary protein and lipid level.Lipid content of fish fed the high-lipid diets(2.31–2.64%) was significantly higher than forthose fed the low-lipid diets (1.51–1.98%)regardless of dietary protein level (Table 4).Ash content (3.49–4.26%) was not signifi-cantly different among treatments. No interac-tion (P > 0.05) between protein and lipid levelon whole body moisture, protein, and ash con-tent of fish were observed, but a significant(P < 0.05) interaction was observed for wholebody lipid content.

Discussion

In the present study, the poor growth perfor-mance of southern flounder juveniles fed the45/10 diet could be as a result of a deficiencyof protein. Insufficient dietary protein reducesprotein assimilation and growth in fish (NRC1993). The BWG for fish fed 45/15 diet wassignificantly higher than those fed 45/10, prob-ably because of the increase in energy level indiet from lipid. In several fish species fed lowprotein diets, using or adding high energy con-taining lipid as a major energy source improved

the growth (Vergara et al. 1996). The presentfinding is consistent with the idea that increas-ing dietary energy by lipid supplementation hasa protein-sparing effect, allowing a more effi-cient use of protein, while reducing N excretionto the environment (Bureau et al. 2002). Pro-tein sparing has been demonstrated in hybrid-striped bass, Morone chrysops × Morone sax-atilis (Nematipour et al. 1992); red sea bream,Pagrus major (Takeuchi et al. 1991); giltheadsea bream, Sparus aurata L. (Vergara et al.1996); turbot, Scophthalmus maximus (Bromley1980); and Atlantic halibut, Hippoglossus hip-poglossus (Aksnes et al. 1996; Helland andGrisdale-Helland 1998).

At a dietary lipid level of 10%, increas-ing protein level from 45 to 50% significantlyincreased BWG. However, further increasingprotein or lipid levels did not affect BWG. Thedietary protein requirement of juvenile south-ern flounder is 50.3% using menhaden mealas protein source and a dietary lipid levelof 14% when reared in a recirculating sys-tem (Gonzalez et al. 2005). Gao et al. (2005)suggested that the optimum protein level indiets for juvenile southern flounder was 51.2%using fish meal and casein–gelatin as proteinsources when reared in a net cage system sup-plied with flow through system. In the presentstudy, increasing lipid levels from 10 to 15% ina 50% protein diet had no significant affect onBWG reared in a recirculating system. Whenfish are fed a diet containing excess lipid,growth may be reduced because of an imbal-ance of digestible energy/CP ratio and excessivefat deposition in the visceral cavity and tis-sues (Lovell 1989). However, increasing lipidlevel from 10 to 15%, did not decrease BWGin the present study. Danielssen and Hjertnes(1993) found no adverse effects of dietary fatlevels up to 22% on growth of juvenile tur-bot. Increasing the lipid level from 8 to 20% inAtlantic halibut diets (Berge and Storebakken1991), from 11 to 21% in Senegalese sole,Solea senegalensis, Kaup diets, (Dias et al.2004), from 3.3 to 16.4% in Japanese flounder,Paralichthys olivaceus diets did not producesignificant differences in weight gain. How-ever, Kim et al. (2004) reported that in juvenile

518 ALAM ET AL.

Table 4. Effects of dietary levels of protein and lipids on body proximate composition (% wet basis) of southern flounder.Values are mean ± SEM (N = 3). Means with different letters in the same column differ significantly (P < 0.05).

Diets (CP/CL) Moisture Protein Lipid Ash

45/10 78.4 ± 0.17 16.0 ± 0.29 1.68 ± 0.16bc 4.26 ± 0.1345/15 77.8 ± 0.21 16.1 ± 0.51 2.54 ± 0.19ab 3.84 ± 0.1750/10 78.8 ± 0.27 16.5 ± 0.23 1.51 ± 0.11c 3.64 ± 0.2850/15 77.8 ± 0.26 16.6 ± 0.38 2.31 ± 0.17ab 3.93 ± 0.2955/10 78.0 ± 0.39 16.4 ± 0.50 1.98 ± 0.18abc 3.79 ± 0.3355/15 78.1 ± 0.08 15.7 ± 0.21 2.64 ± 0.11a 3.49 ± 0.11Com (50/15) 78.1 ± 0.05 16.3 ± 0.15 2.32 ± 0.10ab 3.75 ± 0.66Two-way ANOVA (P -value)Protein (A) 0.0394 0.4221 0.0362 0.9879Lipid (B) 0.0026 0.2469 0.0007 0.1521A × B 0.2907 0.5070 0.0062 0.5933

Japanese flounder, increasing lipid level from6.3 to 12.9% in diets with more than 50% pro-tein caused a significant reduction of growth.We assume that a diet containing 50% proteinand 10% lipid was an appropriate combina-tion for the growth of southern flounder underthe present conditions. These levels were closeto the reported optimum dietary protein andlipid levels for Japanese flounder (Kikuchi andTakeuchi 2002; Kim et al. 2002; Alam et al.2003). This study also confirms that given theoptimum dietary protein the ability to utilizelipid by southern flounder is low, similar towhat was reported in Japanese flounder (Leeet al. 2000a; Alam et al. 2003).

BWG for the juveniles fed the commercialdiet which contained 50% protein and 15%lipid was significantly lower than the formu-lated 50/15 diet. This could be because of thedifferences between these diets in sources ofprotein and lipid and other nutrients such asvitamins, minerals, carbohydrates which are notknown for commercial diet.

In this study, FI for the juveniles was notsignificantly different among the formulateddiets (6.16–7.83 g/fish/42 d) but was signifi-cantly lower for the commercial diet (5.63).This is probably because formulated diets con-tained similar energy levels (14.0 to 15.0 kJ/gdiet) and fish eat to satisfy their energy require-ment (Lee and Putnam 1973). The commercialdiets often included plant meals that give adifferent palatability and texture to the pellets,as well as low-diet digestibility, and this may

have caused the observed lower FI and BWG inthe fish fed commercial diet. The highest FCE(1.29) observed for the 50/10 diet was similarto the reported value for the other flatfish fedfish meal-based diets (Lee et al. 2000b).

PER increased with increasing lipid levelfrom 10 to 15% in the 45% protein diet.However, increasing protein from 50 to 55%caused PER to decrease at both lipid levels,which is consistent with other studies show-ing a decrease in PER with increasing dietaryprotein beyond the optimum level (McGooganand Gatlin 1999; Thoman et al. 1999). A properP/E ratio is important in formulating commer-cial feed. In the present study, dietary P/Eratio ranged from 29.0 to 39.2 mg protein/kJdiet among treatments. The optimum P/E ratiofor the Japanese flounder was 27.5 mg pro-tein/kJ diet as reported by Kim et al. (2004)whereas Lee et al. (2000b) reported 39.9 mgprotein/kJ was optimum for the same species.The optimum P/E ratio for southern flounder inthe present study was 35.4 mg protein/kJ whenfed diet 50/10.

In the present study, to increase energy con-tent in the diets, a higher level of carbohy-drate (wheat starch) was used in the 45 and50% protein diets compared to the 55% proteindiet. The utilization of carbohydrates for south-ern flounder was not evaluated in the presentstudy. However, the growth of flounder in thepresent study was significantly lower for highcarbohydrate-based diets (18% and 13% starchfor the 45/10 and 45/15 diets, respectively)

EFFECT OF DIFFERENT DIETARY PROTEIN AND LIPID LEVELS 519

compared to the other diets containing lowercarbohydrates (0–11% starch). This indicatedthat flounder might not be able to utilize dietarycarbohydrate efficiently as an energy source indiets with imbalanced protein and lipid lev-els. This is consistent with other studies show-ing that lipid rather than carbohydrate is moreefficiently utilized for energy by carnivorousfish (Lee et al. 2002). It is important to deter-mine the utilization of carbohydrate by south-ern flounder, because imbalance in nonproteinenergy sources or levels in the diets might haveadverse influences on growth, nutrient utiliza-tion, and body lipid deposition (Garling andWilson 1976).

There is growing awareness of the effectsof increased body fat deposition on the qual-ity of fish used for human consumption, asthe localization, quantitative importance, andcomposition of body fat depots may affecttheir nutritional value, organoleptic properties,processing yield, and storage stability (Cowey1993). In the present study, the whole bodylipid content of southern flounder (1.51–2.31%wet basis) was well within the range of val-ues for similar size juvenile Japanese floun-der (1.6–1.9%, Hernandez et al. 2005) whenfed a 53% protein and 10% lipid-based diet.Whole body lipid content of southern flounderfed high-lipid diets (2.31–2.64%) was higherthan those fed low lipid diets (1.51–1.98%)regardless of dietary protein levels. This isan agreement with what has been observed inother flatfish such as Japanese flounder (Alamet al. 2003; Kim et al. 2004), Atlantic hal-ibut (Aksnes et al. 1996; Helland and Grisdale-Helland 1998), and Senegalese sole fed highand low lipid diets, (Dias et al. 2004).

Finally, growth of southern flounder fed aformulated diet that contained 50% proteinand 10% lipid was almost double that offish fed a commercial diet which containedthe same dietary protein and lipid level. Thisdemonstrates that the high quality of proteinand lipid sources and a nutritionally balancedformulated diet will be important in developinga commercial diet for maximum growth ofsouthern flounder. The results suggested that acombination of 50% dietary protein and 10%

lipid was optimal for the growth performanceof juvenile southern flounder fed with a fishmeal-based diet reared in a recirculating system.

Acknowledgments

This research was supported by the MAR-BIONC (Marine Biotechnology in North Car-olina) Center for Marine Science, Universityof North Carolina Wilmington and USDA-CSREES. We thank Troy Rezek and Walker D.Wright-Moore for technical assistance.

Literature Cited

Aksnes A., T. Hjertnes, and J. Opstvedt. 1996. Effect ofdietary protein level on growth and carcass compo-sition in Atlantic halibut (Hippoglossus hippoglossusL.), Aquaculture 145:225–233.

Alam, M. S. and W. O. Watanabe. 2005. Southern floun-der. Aqua feeds: formulation and beyond, 2 (4), Fed-ware, Florida, USA, 17–19.

Alam, M. S., S. Teshima, S. Koshio, O. Uyan, andM. Ishikawa. 2003. Effects of dietary protein andlipid levels on growth and body composition of juve-nile Japanese flounder, Paralichthys olivaceus fedintact or crystalline amino acid diets. Journal ofApplied Aquaculture 14(3/4):115–132.

Alam, M. S., S. Teshima, D. Yaniharto, S. Koshio, andM. Ishikawa. 2002. Influence of dietary amino acidpatterns on growth and body composition of juvenileJapanese flounder, Paralichthys olivaceus, Aquaculture210:359–369.

Alam, M. S., W. O. Watanabe, and P. M. Carroll. 2008.Dietary protein requirements of juvenile black sea bassCentropristis striata. Journal of the World AquacultureSociety 39(5):656–663.

AOAC. 1990. [In] K. Helric, editor. Official methodsof analysis of AOAC, 15th edition. Association ofOfficial Analytical Chemists, Inc., Arlington, Texas,USA.

Berge, G. M. and T. Storebakken. 1991. Effect of dietaryfat level on weight gain, digestibility and fillet com-position of Atlantic halibut. Aquaculture 99:331–338.

Berlinsky, D. L., W. King, T. I. J. Smith, R. D. Hamil-ton II, J. Holloway Jr., and C. V. Sullivan. 1996.Induced ovulation of southern flounder Puralichthyslerhosrigma using gonadotropin releasing hormoneanalogue implants. Journal of the World AquacultureSociety 27(2):143–152.

Bromley, P. J. 1980. Effect of dietary protein, lipid andenergy content on the growth of turbot (Scophthalmusmaximus L.). Aquaculture 19:359–369.

Bureau, D. P., S. J. Kaushik, and C. Y. Cho. 2002.Bioenergetics. Pages 1–59 in J. E Halver and R. W

520 ALAM ET AL.

Hardy, editors. Fish nutrition, 3rd edition. AcademicPress, California, USA.

Cho, C. Y. and S. J. Kaushik. 1990. Nutritional energet-ics in fish: Energy and protein utilization in rainbowtrout (Salmo gairdneri ). World Review of Nutrition,Diet 61:162–172.

Cowey, C. B. 1993. Some effects of nutrition on flesh qual-ity of cultured fish. Pages 227–236 in S. J. Kaushikand P. Luquet, editors. Fish nutrition in practice. Pro-ceedings of the IV international symposium fish nutri-tion and feeding, Les Colloques, 61. INRA Editions,Paris, France.

Daniels, H. V. and W. O. Watanabe. 2003. A practicalhatchery manual: production of southern flounder fin-gerlings. Sea Grant Publication, North Carolina, USA.

Danielssen, D. S. and T. Hjertnes. 1993. Effect of dietaryprotein levels in diets for turbot (Scophthalmus max-imus) to market size. Pages 89–96 in S. J. Kaushikand P. Luquet, editors. Fish nutrition in practice. Pro-ceedings of the IV international symposium fish nutri-tion and feeding, Les Colloques, 61. INRA Editions,Paris, France.

De Silver, S. S., R. M. Gunasekera, and K. F. Shim.1991. Interactions of varying dietary protein and lipidlevels in young red tilapia: evidence of protein sparing.Aquaculture 95:305–318.

Dias, J., R. Rueda-Jasso, S. Panserat, L. E. C da Con-ceic, E. F. Gomes, and M. T. Dinis. 2004. Effect ofdietary carbohydrate-to-lipid ratios on growth, lipiddeposition and metabolic hepatic enzymes in juvenileSenegalese sole (Solea senegalensis . Kaup Aquacul-ture Research 2004 35:1122–1130.

Gao, Y., J. LV, and Q. Lin. 2005. Effects of proteinlevels on growth, feed utilization, nitrogen and energybudget in juvenile southern flounder, Paralichthyslethostigma. Aquaculture Nutrition 11:427–433.

Garling, D. L. and R. P. Wilson. 1976. The optimumprotein-to-energy ratio for channel catfish, Ictaluruspunctatus . Journal of Nutrition 106:1368–1375.

Gonzalez, S., S. R. Craig, E. McLean, M. H. Schwarz,and G. J. Flick. 2005. Dietary protein requirement forsouthern flounder, Paralichthys lethostigma. Journal ofApplied Aquaculture 17(3):37–50.

Helland, S. J. and B. Grisdale-Helland. 1998. Growth,feed utilization and body composition of juvenileAtlantic halibut (Hippoglossus hippoglossus) fed dietsdiffering in the ratio between the macronutrients.Aquaculture 166:49–56.

Hernandez, L. H. H., S. Teshima, S. Koshio, M. S.Alam, M. Ishikawa, and H. Tanaka. 2005. Dietaryvitamin A requirement of juvenile Japanese flounder,Paralichthys olivaceus . Aquaculture Nutrition 11:3–9.

Kikuchi, K. and T. Takeuchi. 2002. Japanese flounder,Paralichthys olivaceus . Pages 113–120 in C. D.Webster and C. E. Lim, editors. Nutrient requirementsand feeding of finfish for aquaculture. CABI publish-ing, Cambridge, MA, USA.

Kim, K. W., X. J. Wang, and S. C. Bai. 2002. Optimumdietary protein level for maximum growth of juvenile

olive flounder Paralichthys olivaceus (Temminck andSchlegel). Aquaculture Research 33:673–679.

Kim, K. W., X. Wang, S. M. Choi, G. J. Park, andS. C. Bai. 2004. Evaluation of optimum dietaryprotein-to-energy ratio in juvenile olive flounder Par-alichthys olivaceus (Temminck et Schlegel) Aquacul-ture Research 35:250–255.

Kramer, C. Y. 1956. Extension of multiple range teststo group means with unequal number of replications.Biometrics 12:307–310.

Lee, S. M., S. H. Cho, and K. D. Kim. 2000a. Effectsof feeding frequency and dietary energy level ongrowth and body composition of juvenile flounder,Paralichthys olivaceus (Temminck & Schlegel). Aqua-culture Research 31:917–921.

Lee, S. M., S. H. Cho, and K. D. Kim. 2000b. Effectsof dietary protein and energy levels on growth andbody composition of juvenile flounder (Paralichthysolivaceus). Journal of the World Aquaculture Society31:306–315.

Lee, S.-M., I. G. Jeon, and J. Y. Lee 2002. Effects ofdigestible protein and lipid levels in practical feedon growth, protein utilization and body compositionof juvenile rockfish (Sebastes schlegeli ). Aquaculture211:227–239.

Lee, D. J. and G. B. Putnam. 1973. The response ofrainbow trout to varying protein/energy ratios in a testdiet. Journal of Nutrition 103:916–922.

Lovell, R. T. 1989. Nutrition and feeding of fish. VanNostrand Reinhold, New York, New York, USA.

McGoogan, B. B. and D. M. Gatlin. 1999. Dietarymanipulations affecting growth and nitrogenous wasteproduction of red drum, Sciaenops ocellatus I. Effectsof dietary protein and energy levels. Aquaculture178:333–348.

Nematipour, G. R., M. L. Brown, and D. M. Gatlin III.1992. Effects of dietary energy: protein ratio on growthcharacteristics and body composition of hybrid stripedbass, Morone chrysops’ M. sax-atilis. Aquaculture107:359–368.

NRC (National Research Council). 1993. Nutrientrequirements fishes. National Academy Press, Wash-ington DC, USA.

Takeuchi, T., Y. Shiina, and T. Watanabe. 1991. Suitableprotein and lipid levels in diet for fingerlings of redsea bream Pagrus major . Nippon Suisan Gakkaishi55(2):521–527.

Thoman, E. S., D. A. Davis, and C. R. Arnold. 1999.Evaluation of growout diets with varying protein andenergy levels for red drum (Sciaenops ocellatus).Aquaculture 176:343–353.

Vergara, J. M., L. Robaina, M. de Izquierdo, and M. laHiguera. 1996. Protein sparing effect of lipids in dietsfor fingerlings of gilthead sea bream. Fisheries Science62(4):624–628.

Watanabe, W. O., P. M. Carroll, and H. V. Daniels.2001. Sustained, natural spawning of southernflounder Paralichthys lethostigma under an extended

EFFECT OF DIFFERENT DIETARY PROTEIN AND LIPID LEVELS 521

photothermal regime. Journal of the World Aquacul-ture Society 32:153–166.

Watanabe, W. O., C. A. Woolridge, and H. V. Daniels.2006. Progress toward year round spawning ofsouthern flounder broodstock by manipulation ofphotoperiod and temperature. Journal of the WorldAquaculture Society 37: 256–272.

Wenner, C. A., W. A. Roumillat, J. E. Moran Jr., M.B. Maddox, L. B. Daniel, and J. W. Smith. 1990.Investigation on the life history and population dynam-ics of marine recreation fishes in South Carolina: PartI. South Carolina Wildlife and Marine DepartmentFinal Report Project F-37. Charleston, South Carolina,USA.